Laura Lundquist

(Missoula Current) When grizzly bears venture outside of designated recovery areas, they face more of a threat from human interaction. A new map of their likely migration routes could guide state efforts to minimize conflict within those corridors.

Building upon initial modeling efforts in 2017, a new computer model can more accurately predict how grizzly bears would likely move not only between the Northern Continental Divide and greater Yellowstone recovery areas, but also between the Continental Divide, Cabinet-Yaak and Bitterroot recovery areas.

Sarah Sells, ecologist with the U.S. Geological Survey Cooperative Wildlife Research Unit at the University of Montana, led the study that was published last week in the journal Biological Conservation.

“We wanted to do this work because the U.S. Fish and Wildlife Service and Montana Fish, Wildlife & Parks identified that further mapping of the connectivity pathways was one of the most important research needs for grizzly bear conservation,” Sells said.

Wildlife managers started paying more attention to connectivity in March 2016 when the U.S. Fish and Wildlife Service announced that it intended to delist the Greater Yellowstone population of grizzly bears. Some bear advocates were concerned that delisting might reduce efforts to conserve grizzly habitat along migration corridors outside the Greater Yellowstone area. They pointed out that the Cabinet-Yaak and Bitterroot populations are heavily dependent on bear migration.

In 2017, USGS biologist Christopher Peck with the Interagency Grizzly Bear Study Team published research that used a computer model based on the movements of collared grizzly bears to predict the likely migration corridors between the Northern Continental Divide and Greater Yellowstone recovery areas.

The resulting maps highlighted a lattice of likely migration routes with smears of dark purple that tended to hug the mountain ranges between the Bob Marshall Wilderness Complex and Yellowstone Park, avoiding large valleys and population centers.

Fast forward to now, and Sells’ computer model produces what appears to be similar smears across the landscape, although hers are in blue. Sells said that although the main routes are the same, her map provides more nuanced information, because she used different data and gave grizzlies a slightly different outlook.

“But I think it’s a good thing when repeated science keeps saying, yeah, this is pretty much the same result,” Sells said.

Grizzly bear locations between established recovery "zones."
Grizzly bear locations between established recovery "zones."

First, where Peck considered only male bears - males tend to be the ones who roam farther from their home territory - Sells used the data from 46 female and 19 male collared bears. Some people might remember Ethyl, the female grizzly who in 2014 wandered 2,800 miles between Canada, Montana and northern Idaho, somehow safely crossing Interstate 90 a handful of times.

Second, where Peck took a population-level approach, lumping all the bear data together and building his model off that, Sells took her model down to the level of individual bears.

“We fit an individual model to each bear, because we know that bears have a lot of individual variation in behavior - we wanted to be sure to capture all of that. So we ran these simulations a whole bunch of times for a whole bunch of different bears,” Sells said. “The (two) maps, at first glance, look similar between the NCDE and the Greater Yellowstone, but ours is a lot more nuanced. We don’t show this long, wide passageway over a highway, for example. We have more nuance there, showing this little section of road is good and this little section is good. I think it will help with implementing some of the conservation measures that we could put in like passages at roads to protect humans and bears.”

Sells also considered the bears’ intent, or lack thereof, by allowing their movement to be undirected. In his model, Peck constrained bear movement with start and end points, assuming the bear would leave one recovery area and move to the other.

“That assumes a bear will know where it’s going- that’s not necessarily realistic. We provided only a start point and said, ‘Show us where you will go,’” Sells said. “That will be helpful in making decisions outside the recovery zones.”

With the state of Montana petitioning the U.S. Fish and Wildlife Service for delisting of the Northern Continental Divide population, identifying and preserving migration corridors outside the recovery areas is essential for the future of the bear.

In 2017, Peck acknowledged that “successful immigration events will likely remain rare given the current distance between the populations. If the two populations continue to expand and this distance decreases, the likelihood of successful immigration will increase accordingly.”

The distance can decrease only if bears can survive outside the recovery areas. The U.S. Fish and Wildlife Service must guarantee that to attempt another delisting. In his 2018 ruling negating the delisting of the Yellowstone population, federal district court judge Dana Christensen required the Service to show there was connectivity between populations to prevent inbreeding.

Roads are a major barrier to migration, with a few grizzlies killed almost every year in vehicle collisions. In 2020 and 2021, one grizzly bear dubbed “Lingenpolter” spent weeks searching along the north edge of Interstate 90 for a way across only to retreat repeatedly. Wildlife crossing structures could provide safe passage, and Sells’ maps show some of the best places to focus on building crossing structures.

“We’re using it to begin discussions about some of the road work between Florence and Lolo. That’s where one of the pathways over to the Bitterroot (recovery area) really concentrates,” Sells said. “The Ninepipe Refuge - there was a discussion earlier this summer about getting some mitigation in that area, because it’s also where our maps show another strong corridor.”

Contact reporter Laura Lundquist at

More From Missoula Current